001052293 001__ 1052293
001052293 005__ 20260122203309.0
001052293 0247_ $$2doi$$a10.1016/j.mlwa.2025.100790
001052293 0247_ $$2datacite_doi$$a10.34734/FZJ-2026-00910
001052293 037__ $$aFZJ-2026-00910
001052293 082__ $$a004
001052293 1001_ $$00000-0001-9915-3039$$aSardhara, Trushal$$b0
001052293 245__ $$aDeep learning-based 3D reconstruction of dentate nuclei in Friedreich’s ataxia from T2*weighted MR images
001052293 260__ $$aAmsterdam$$bElsevier$$c2025
001052293 3367_ $$2DRIVER$$aarticle
001052293 3367_ $$2DataCite$$aOutput Types/Journal article
001052293 3367_ $$0PUB:(DE-HGF)16$$2PUB:(DE-HGF)$$aJournal Article$$bjournal$$mjournal$$s1769088056_20964
001052293 3367_ $$2BibTeX$$aARTICLE
001052293 3367_ $$2ORCID$$aJOURNAL_ARTICLE
001052293 3367_ $$00$$2EndNote$$aJournal Article
001052293 520__ $$aDentate nucleus (DN) degeneration is a key neuropathological feature in Friedreich’s ataxia (FRDA), and its accurate quantification is critical for understanding disease progression. However, its visualization and volumetry require iron-sensitive imaging techniques and time-consuming segmentation procedures, posing challenges for conventional ML approaches due to small datasets typical of rare diseases. We present a transfer learning–based machine learning pipeline for automated DN segmentation that directly uses standard T2*-weighted Magnetic Resonance Imaging (MRI), which highlights the DN without additional processing, and is designed to perform robustly with limited annotated data. Using 38 manually labeled subjects (18 FRDA, 20 controls), the model was validated via five-fold cross-validation and an independent hold-out test set, achieving Dice scores of 0.81–0.87 and outperforming classical atlas-based methods. Pretraining improved performance by ∼10% in patients and >5% in controls. Applied to 181 longitudinal scans from 33 FRDA patients and 33 controls, the model revealed significantly reduced DN volumes in FRDA, with reductions correlating with disease duration and clinical severity over time. Our approach provides a scalable and reproducible segmentation framework, requiring minimal annotated data and no preprocessing, while demonstrating robust performance across cross-validation and independent testing. Additionally, it enables the first longitudinal volumetric analysis of DN in FRDA using standard T2*-weighted MRI, demonstrating its practical utility for monitoring neurodegenerative changes. Overall, this work illustrates how transfer learning can overcome data scarcity in rare diseases and provides a robust methodology for automated MRI segmentation in both research and clinical applications.
001052293 536__ $$0G:(DE-HGF)POF4-5251$$a5251 - Multilevel Brain Organization and Variability (POF4-525)$$cPOF4-525$$fPOF IV$$x0
001052293 588__ $$aDataset connected to CrossRef, Journals: juser.fz-juelich.de
001052293 7001_ $$0P:(DE-HGF)0$$aDadsena, Ravi$$b1
001052293 7001_ $$0P:(DE-HGF)0$$aAydin, Roland C.$$b2
001052293 7001_ $$aHilgers, Ralf-Dieter$$b3
001052293 7001_ $$aHorn, Leon$$b4
001052293 7001_ $$0P:(DE-Juel1)171786$$aSchulz, Jörg B.$$b5
001052293 7001_ $$0P:(DE-Juel1)177889$$aReetz, Kathrin$$b6$$eCorresponding author
001052293 7001_ $$aRomanzetti, Sandro$$b7
001052293 7001_ $$aDogan, Imis$$b8
001052293 7001_ $$aLischewski, Stella A.$$b9
001052293 7001_ $$0P:(DE-Juel1)174172$$aKonrad, Kerstin$$b10
001052293 7001_ $$aPishnamaz, Miguel$$b11
001052293 7001_ $$aPraster, Maximillian$$b12
001052293 7001_ $$aClavel, Thomas$$b13
001052293 7001_ $$aJankowski, Vera$$b14
001052293 7001_ $$aJankowski, Joachim$$b15
001052293 7001_ $$aPabst, Oliver$$b16
001052293 7001_ $$aMarx-Schütt, Katharina$$b17
001052293 7001_ $$aMarx, Nikolaus$$b18
001052293 7001_ $$aMöllmann, Julia$$b19
001052293 7001_ $$aJacobsen, Malte$$b20
001052293 7001_ $$0P:(DE-Juel1)177772$$aDukart, Juergen$$b21
001052293 7001_ $$0P:(DE-Juel1)131678$$aEickhoff, Simon$$b22
001052293 773__ $$0PERI:(DE-600)3052291-2$$a10.1016/j.mlwa.2025.100790$$gVol. 22, p. 100790 -$$p100790 -$$tMachine learning with applications$$v22$$x2666-8270$$y2025
001052293 8564_ $$uhttps://juser.fz-juelich.de/record/1052293/files/1-s2.0-S2666827025001732-main.pdf$$yOpenAccess
001052293 909CO $$ooai:juser.fz-juelich.de:1052293$$popenaire$$popen_access$$pVDB$$pdriver$$pdnbdelivery
001052293 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)171786$$aForschungszentrum Jülich$$b5$$kFZJ
001052293 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)177889$$aForschungszentrum Jülich$$b6$$kFZJ
001052293 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)174172$$aForschungszentrum Jülich$$b10$$kFZJ
001052293 9101_ $$0I:(DE-588b)5008462-8$$6P:(DE-Juel1)131678$$aForschungszentrum Jülich$$b22$$kFZJ
001052293 9131_ $$0G:(DE-HGF)POF4-525$$1G:(DE-HGF)POF4-520$$2G:(DE-HGF)POF4-500$$3G:(DE-HGF)POF4$$4G:(DE-HGF)POF$$9G:(DE-HGF)POF4-5251$$aDE-HGF$$bKey Technologies$$lNatural, Artificial and Cognitive Information Processing$$vDecoding Brain Organization and Dysfunction$$x0
001052293 915__ $$0StatID:(DE-HGF)0150$$2StatID$$aDBCoverage$$bWeb of Science Core Collection$$d2025-01-07
001052293 915__ $$0LIC:(DE-HGF)CCBY4$$2HGFVOC$$aCreative Commons Attribution CC BY 4.0
001052293 915__ $$0StatID:(DE-HGF)0112$$2StatID$$aWoS$$bEmerging Sources Citation Index$$d2025-01-07
001052293 915__ $$0StatID:(DE-HGF)0501$$2StatID$$aDBCoverage$$bDOAJ Seal$$d2021-01-29T11:25:19Z
001052293 915__ $$0StatID:(DE-HGF)0500$$2StatID$$aDBCoverage$$bDOAJ$$d2021-01-29T11:25:19Z
001052293 915__ $$0StatID:(DE-HGF)0700$$2StatID$$aFees$$d2025-01-07
001052293 915__ $$0StatID:(DE-HGF)0510$$2StatID$$aOpenAccess
001052293 915__ $$0StatID:(DE-HGF)0030$$2StatID$$aPeer Review$$bDOAJ : Anonymous peer review$$d2021-01-29T11:25:19Z
001052293 915__ $$0StatID:(DE-HGF)0561$$2StatID$$aArticle Processing Charges$$d2025-01-07
001052293 915__ $$0StatID:(DE-HGF)0300$$2StatID$$aDBCoverage$$bMedline$$d2025-01-07
001052293 915__ $$0StatID:(DE-HGF)0199$$2StatID$$aDBCoverage$$bClarivate Analytics Master Journal List$$d2025-01-07
001052293 920__ $$lyes
001052293 9201_ $$0I:(DE-Juel1)INM-7-20090406$$kINM-7$$lGehirn & Verhalten$$x0
001052293 980__ $$ajournal
001052293 980__ $$aVDB
001052293 980__ $$aUNRESTRICTED
001052293 980__ $$aI:(DE-Juel1)INM-7-20090406
001052293 9801_ $$aFullTexts